Alexa and red mx - comparative test

Very intriguing test but the formatting of that article makes it kind of hard to follow.

Noah

Unlock the secrets of 24p, HD and Final Cut Studio with Call Box Training. Featuring the Canon 5D Mark II and 7D.

Nice test.
We all know the Alexa has more DR then the current MX based Red One and Epic but it would be nice to see a comparison on actual usability. Even something as simple as booting up and turning on can be a bit of a challenge for some cameras.

This is just a comment, because I have seen this been discussed at several fora:

Hi and thanks for the interest in my test!

As to development:
All Arri material was shot in logC, all RED material was developed with redlogFilm.

I am considering writing a separate article on how to get all DR off the RED... And I have been discussing that at length on several fora. I am not unaware of the difference seen on set and what one can get in post...

As to choice of ASA/ISO
I use some space to documet Arris claim that all DR is available independently of ISO setting between 200 and 1600 when shooting logC

Also I note that from 800 on the Alexa, analogue gain kicks in.

On the RED side...
REDs advice is shooting 800, and actually I do fully agree. You get 1,25 stop highlightprotection shooting that way. Problem is, that if you don't set the ISO back to 320 (or lower) in post (even with redlogFilm), you will push some of the highlights into clipping. You do have to lower the ISO and compensate in the lower region with a curve to get the full latitude off the MX.

Ideally we should see the highlights from the 320 or setting and the lowlights from the 800+ settings, but unfortunately, that is not available. It is in the RAW files, though...

I considered this explanation a bit confusing for this comparison, and as the intention when matching the cameras were basically to "fill both to the most" both in the top and the bottom, I chose to shoot at 320 on both to avoid a lot of "voodoo confusion".

As I have shown on the previous pages, the Alexa do not lose any DR this way.
On the RED side, I could use the "view RAW" option to determine when the highlights clipped on the waveform, and do the rest in post.
BTW, the waveform and the redlogFilm file were totally in agreement.

If I had chosen to shoot the RED at 800, I would have had to spend a lot of text to explain why I overexposed the RED by 1,25 stop to fill up the latitude. Or... effectively choke the RED for 1,25 stop in the top.
Exposing at 320 made that comparison simpler to understand for most readers, it seems.

I have one test here, showing off a bit on how this works, without directly discussing it: Unfortunately RED decided to stop the discussion over at reduser, but I try to follow up wherever I see answers.

http://www.gunleik.com/tests/tungbrack


Cheers

Gunleik

http://www.gunleik.com/tests/